Effect of documented and undocumented psychiatric conditions on length of stay and discharge destination after total knee arthroplasty.

Introduction: Length of stay (LOS) and readmissions are quality metrics linked to physician payments and substantially impact the cost of care. This study aims to evaluate the effect of documented and undocumented psychiatric conditions on LOS, discharge location, and readmission following total knee arthroplasty (TKA).

Methods: Retrospective review of all primary, unilateral TKA from 2015 to 2020 at a high-volume, academic orthopedic hospital was conducted. Patients were separated into three cohorts: patients with a documented psychiatric diagnosis (+Dx), patients without a documented psychiatric diagnosis but with an actively prescribed psychiatric medication (-Dx), and patients without a psychiatric diagnosis or medication (control). Patient demographics, LOS, discharge location, and 90 days readmissions were assessed.

Results: A total of 2935 patients were included; 1051 patients had no recorded psychiatric medications (control); 1884 patients took at least one psychiatric medication, of which 1161 (61.6%) were in the-Dx and 723 (38.4%) were in the +Dx cohort. Operative time (+Dx, 103.4 ± 29.1 and -Dx, 103.1 ± 28.5 vs. 93.6 ± 26.2 min, p < 0.001 for both comparisons) and hospital LOS stay (+ Dx, 3.00 ± 1.70 and -Dx, 3.01 ± 1.83 vs. 2.82 ± 1.40 days, p = 0.021 and p = 0.006, respectively) were greater for patients taking psychiatric medications when compared to the control group. Patients taking psychiatric medication with or without associated diagnosis were significantly more likely to be discharged to a secondary facility-22.8% and 20.9%, respectively-compared to controls, at 12.5% (p < 0.001). Ninety-day readmission rates did not differ between the control and both psychiatric groups (p = 0.693 and p = 0.432, respectively).

Conclusion: TKA patients taking psychiatric medications with or without a documented psychiatric diagnosis have increased hospital LOS and higher chances of discharge to a secondary facility. Most patients taking psychiatric medication also had no associated diagnosis. Payment models should consider the presence of undocumented psychiatric diagnoses when constructing metrics. Surgeons and institutions should also direct their attention to identifying, recording, and managing these patients to improve outcomes.

Level Iii Evidence: Retrospective Cohort Study.